Sunteți pe pagina 1din 14

Vol.

II, Issue 4
Scientific Papers (www.scientificpapers.org)
July 2013
Journal of Business Management and Applied Economics

Electricity Consumption and Economic Growth in Nigeria

Authors: Adeyemi A. Ogundipe, Department of Economics and Development Studies,


Covenant University, Ota, Ade.ogundipe@covenantuniversity.edu.ng,
yemi_keke2001@yahoo.com, Ayomide Apata, Department of Economics, University of
Dundee, Dundee, United Kingdom, ayomideapata@yahoo.co.uk

The paper seeks to examine the relationship between electricity consumption and economic growth in
Nigeria using the Johansen and Juselius Co-integration technique based on the Cobb-Douglas growth
model covering the period 1980-2008. The study adopted also conducted the Vector Error Correction
Modelling and the Pairwise Granger Causality test in order to empirically ascertain the error correction
adjustment and direction of causality between electricity consumption and economic growth. The study
found the existence of a unique co-integrating relationship among the variables in the model with the
indicator of electricity consumption impacting significantly on growth. Also, the study shows an evidence
of bi-directional causal relationship between electricity consumption and economic growth. Prominent
among the policy recommendation, is the need to strengthen the effectiveness of energy generating
agencies by ensuring periodic replacement of worn-out equipment in order to drastically curtail
transmission power losses.

Keywords: Electricity Consumption, Economic Growth, Co-integration, and Causality test

Introduction countries, (Morimoto and Hope 2001). As sited


in Morimoto and Hope 2001; Ferguson et al
Poor access to electricity in Nigeria has been a (2000) study of the correlation between
major impediment to Nigerias economic electricity use and economic growth in Sri
growth. SMEs have been adjudged as the engine Lanka found a very high positive correlation of
of economic growth but its performance is 0.993, thereby concluding the existence of
grossly dismal due to inadequate power supply. strong correlation between electricity use and
Researchers have identified the increase in economic development. Increasing incidence of
energy use as a vital component of emerging power shortages has been identified as
economies; economic growth of the South Asia responsible for the dwindling growth of most
Association for Regional Cooperation (SAARC) underdeveloped countries and this is not
countries involving Bangladesh, India, unconnected with the inabilities to develop new
Pakistan and Sri Lanka is closely related to its generating capacity as hydropower has been the
energy consumption which is an impediment only source of power, thereby diminishing
for enhancing export values, increasing electricity supply severely during droughts
remittances receipts from manpower supply, (Ferguson et al 2000). In order to ensure an
Sheriff (2002). Whether African economies, appropriate recovery of the socio-economic
most especially Nigeria are ready for process of Nigeria within the framework of
developmental take-off should be based on its effective economic system, development,
readiness to ensure adequate and regular power enhancing structures, patterns and evolution of
supply, which represent a crucial factor that production, allocation and utilization of its vast
supports economic growth in developing

1
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

resources, similarly ensuring optimal others (wood fuel and solar) are used in their
development and efficient management of crude forms for heating, cooking and lighting.
available resources, equitable allocation of such The responsibility of production and
resources and effective utilization in order to distribution of electricity was saddled with the
ultimately achieve economic development; the National Electric Power Authority (NEPA),
issue of electricity (power) availability needs to established by decree no. 24 of 1972 until
be taken as a vocal point in development recently when the sector was deregulated in
planning, that is, the modern technologies order to allow private participation. The NEPA
needed to drive economic development are was charged with the statutory monopoly
strictly tied to the use of energy. This therefore, power to over-see electricity development
is a function of adequate supply and throughout the country and produce electricity
distribution of energy, most especially under a high proportion of in-operational
electricity. This study therefore becomes generating plant capacities of 27%, overloaded
imperative in analyzing the challenges of and overstretched transmission lines; in
electricity supply and to examine the level of addition, the problem of hydrological
electricity induced growth in the Nigerian inadequacies in hydro-electric plants especially
economy. The study is outlined into the within the period of dry season. The foregoing
following sections; section two focused on the challenges coupled with illegal access to
background information/stylized facts on the transmission lines have culminated into
subject matter in Nigeria; section three, briefly frequent breakdown of electricity equipment
link the incidence of energy consumption and (seemingly due to overload) and a large
growth in line to the existing literatures; section quantum of electricity losses in the
four provides the theoretical framework and transmission system (ranging between 20-30%),
model formation for the study and section five NEPA often responded to these anomalies by
concludes with policy implementation. creating an electricity supply-demand artificial
balance such as rationing, shedding and
Stylized/background facts suppressed demand services; all these have
resulted in the low quantum of electricity
Evidences have shown that Nigeria is primarily available for consumption. This current status
an energy store house accommodating of electricity supply in Nigeria reflects a
resources such as coal and lignite, natural gas, situation of supply crisis in which industrial
crude oil, solar, hydro, nuclear, wood fuel, growth and socio-economic development paces
geothermal, tide, biogas and biomass. In spite are kept below the potential of the economy
of the available vast resources, only four sources (Ayodele, 2000; FRN 1975; WORLD BANK 1991;
(coal, crude oil, natural gas and hydro) are Ayodele, 1992 & 1999).
currently utilized in processed forms while two

Energy consumption
120000
100000
80000
60000
EGU
40000
20000
0
1970 1980 1990 2000 2010 2020

Figure 1: Energy Consumption Trend in Nigeria

Source: computed from World Development Indicator Database

2
Vol. II, Issue 4
Scientific Papers (www.scientificpapers.org)
July 2013
Journal of Business Management and Applied Economics

In line with the trend witnessed in likewise the energy use per capita has steadily
most developing countries, Nigeria energy been rising until 2005 where a decline was
consumption has increasingly experienced an witnessed and afterwards has been steadily
upward trend with over 23% increase in energy increasing. The continuous increase in energy
use between 2000 and 2008 (see figure 1). Since consumption is quite consistent with GDP but
1970, Nigerias energy consumption has the energy consumption has been increasing at
consistently maintained an upward trend, a faster rate than GDP (see figure 2).

30
25
20 Energy
15 Consumption
10 GDP
5
0
1970 1980 1990 2000 2010 2020

Figure 2: Log Trend Pattern of Energy Consumption and GDP

In the face of the raging need for situation but describes an electricity supply
energy consumption, distribution losses (see crisis has activated wide spread poverty in
table below) and the NEPA devices to allocate Nigeria as the businesses of the middle class
available electricity to consumers; it is therefore populace has been eradicated due to increasing
evident that the quantum of electricity does not energy cost and multi-nationals have sort
meet the actual demand for electricity. The greener pasture in neighboring countries.

6E+10

5E+10

4E+10
Losses
3E+10
Production
2E+10 Consumption

1E+10

0
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29

Figure 3: Total Electricity Production, Consumption and Losses (1980-2009)

Source: Computed from World Development Indicators Database

The issue of power losses has been a of the electricity generated are unavoidably lost
major challenge for the electricity generating in transmission process, several power plants
agencies in Nigeria, majority of this problem is have been erected in the country from the
due to vandalism and inadequate and worn-out inception of democratic governance in 1999 but
electricity transmission equipments. Over 45% the Nigeria economy is yet to appropriate the

3
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

benefits of the huge investment, as electricity successfully wiped off cottage industries due to
unavailability still remains an invisible ghost high cost of generating power independently.
haunting the nations economy and has

Table 1: Electricity Generation and Consumption

year installed total total consumption power losses in Losses (million


capacity(mw) generation (million kwh) transmission kwh % of total)
(million kwh)
1990 4,548 13,462.9 7,870.5 5,592.4 41.5
1991 4,548 14,166.6 8,292.0 5,874.6 41.5
1992 4,548 14,833.8 8,699.0 6,134.8 41.4
1993 4,586.6 14,504.6 9,998.3 4,506.3 31.1
1994 4,548.6 15,531.6 9,593.9 5,937.1 38.2
1995 4,548.6 15,856.6 9,435.9 6,420.7 40.5
1996 4,548.6 16,242.8 9,051.8 7,191.0 44.3
1997 4,548.6 16,116.8 8,843.2 7,273.7 45.1
1998 5,400.0 15,110.0 8,521.2 6,588.8 43.6
1999 5,876.0 16,088.7 8,576.3 7,512.4 46.7

Source: Compiled by author from CBN, 2002

Brief Review of Literature


Hayat F.M, Hamed N and Inqman M (2012)
investigated the relationship between energy
Economic debates surrounding the research consumption and economic growth in Pakistan
cant explicitly link the relationship between for the period of 1973-2006 and found a positive
energy consumption and economic growth to relationship with a unidirectional causality
theories, though empirical evidences have from GDP to energy consumption. A similar
stated results for about two decades. The study of Kouakou A. K (2010) in Cote dIvoire
seminal work of Kraft and Kraft (1978) covering 1971-2008 found a bi-directional
presented the premier study on the causal causality between per capita electricity
relationship between economic growth and consumption and per capita GDP. A study by
energy consumption; Also, research evidences Ouadraogo N.S (2012) for fifteen countries of
have discovered a story correlation between ECOWAS from 1980-2008 using a panel co-
electricity use and wealth creation (Ghosh 2002; integration technique found GDP and energy
Shiu and Lam 2004; Morimoto and Hope 2004; consumption as well as GDP and electricity to
Jumbe 2004; Wolde-Rufael 2004; Narayan and exhibit a long-run co-integrating relationship,
Smyth 2005; Yoo 2005. Altinay and Karagol likewise found a unidirectional causality
(2004) discovered a rising energy need for most running from GDP to energy consumption.
developing countries; Turkey also facing an ever Ciarreta A. and Zarraga A (2007) using a
increasing electricity demand experienced 8.1% standard Granger causality test in a VAR found
per annum in the average growth rates of total a unidirectional linear causality running from
electricity consumption between 1980 and 2000; real GDP to electricity. Also, a premier work
Nigeria also face similar trend experiencing from by Morimoto R and Hope C found
about 23% increase in energy use between 2000 electricity supply to have a significant impact
and 2008. on variation in GDP in Sri Lanka; the result
Several studies have attempted the obtained is similar to Yang (2000). Several
relationship and direction of causality between studies, most especially in developing
energy consumption and economic growth, economies have found electricity consumption
Ahmed N, to be a significant determinant of GDP growth

4
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

(Soytas and Sari 2003; Asafu-adjaye 2000; represents Gross Domestic product,
Ferguson et al 2000; Altinay G and Karagol E is the electricity consumption (Kilowatt per
2005). Contrarily to the forgoing assumptions, hour), is total labor force, is the stock
Mehrara M. and Musai M (2002) using a panel of capital and is the white noise term. The a
analysis of 11 selected oil exporting countries priori expectation is such that 1 , 2 , 3 > 0.
found that electricity use does not have any The equation for Granger causality test can
significant effects on GDP. be specified as follows:

= 1
Methodological Framework
+ 1 + 1
The Model
= 1
The study adopts a Cobb-Douglas production
function with constant returns to scale similar + 1 + 1
to Ahmed N et al (2012).
Data Sources and Measurement
=
In the model above, Y is the total The data used in the study is drawn from the
production (output), L is the labor input, K is
World Development indicators and central
capital input and A is the total factor
productivity. and are the output elasticitys bank of Nigeria statistical Bulletin. Data for
of labor and capital respectively. gross fixed capita formation, labor force and
= 0 1 2 3
energy use are sourced from the World
The explicit form of the model stated in
a log linearized form can presented as follow: Development Indicators of World bank, 2012
= 0 + 1 + 2 while gross fixed capita formation is drawn
+ 3 +
from the central bank of Nigeria statistical
Bulletin, 2010.

Variable Description Source measurement


lgdp Gross Domestic Product World Development Indicators of Constant 2000 US$
World Bank, 2012
lkap Gross Fixed Capital Central Bank of Nigeria Statistical Constant 2000 US$
Formation Bulletin, 2010
llab Labour force World Development Indicators of Number
World Bank, 2012
elec Electricity Consumption World Development Indicators of KwH
World Bank, 2012

Source: Compiled by author

Econometric Analysis series property of gross domestic product,


capital stock, labor and energy consumption in
This aspect attempts an empirical investigation
order to avoid the occurrence of a spurious
of the effect of energy consumption (proxy for
regression. Determining the order of
electricity consumption) on gross domestic
integration of the variables involves subjecting
product. The section starts with examining the
the data series to a unit root testing; here two
time series characteristics of the variables
unit root test procedure shall be adopted-the
included in the model; that is, testing the time
Augmented Dickey Fuller (ADF) and the Philip

5
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

Perron (PP) test. After ascertaining the order of Unit Root Testing
integration, we can then proceed to estimating
The section examines the unit root property of
the Johansen and Joselius co-integration
the variables in the model using ADF and PP
analysis in order to test for the existence of a
test with the inclusion of trend and intercepts
co-integrating relationship among the variables.
components in the test equations at both levels
Finally, a test of causal relationship between
and first difference. All the variables appear to
energy consumption and GDP is conducted
be stationary at first difference at 5%
using a pairwise granger causality test.
significance level.

Variable Level First difference


ADF PP ADF PP
lgdp -3.7157** -3.4144** -5.3803* -5.4849*
lkap -1.0143 -1.0143 -4.3094** -4.3307**
llab -2.6504 -2.7639 -5.5434* -5.5157*
Lelec -1.0559 -0.8049 -8.0410* -8.1822*

Source: Computed by author using e-views 5.0


*significance at 1% **significance at 5% ***significance at 10%

Johansen Co-integration Test

The study proceeds to test for the existence of compare unrestricted co-integration rank test
co-integration among the variables in the available from the trace and maximum
model; this is based on the representation of eigenvalue test with the corresponding critical
the approach specified by Johansen and Juselius values due to Mackinnon-Haug-Michelis (1999).
(1990). The Johansen test for co-integration The result indicates that the trace statistic show
provides an analytical statistical framework for an evidence of a unique co-integration
ascertaining the long-run relationship between equation, which implies an existence of long
the economic variable (Agbola, 2004). The table run equilibrium relationship among the
observed variables.

Trace Test
eigenvalue Trace Static critical value at 0.05 Prob Hypothesized no. of
CE(s)
0.975426 97.51483 55.24578 0.0000 None*
0.667875 34.51174 35.01090 0.0565 At most 1
0.469386 15.77360 18.39771 0.1122 At most 2
0.254827 5.000350 3.841466 0.0253 At most 3
Maximum Eigenvalue
0.975426 63.00309 30.81507 0.0000 None*
0.667875 18.73815 24.25202 0.2267 At most 1
0.469386 10.77325 17.14769 0.3299 At most 2
0.254827 5.000350 3.841466 0.0253 At most 3
Co-integration co-efficient normalized on growth
LGDP LKAP LLAB LELEC
1.000000 -0.073886 3.324912 -0.125961
(0.00235) (0.11649) (0.00723)

Source: Computed by author using e-views 5.0

6
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

The normalized energy induced growth


equation shows the respective effect of the Vector Error Correction Model
explanatory variables on the regressand. The
explanatory variables all exert an inelastic and The table below indicates that estimated lagged
significant impact on the explained variable, error correction term of growth. The magnitude
except labour force which exerts an elastic of the error correction term is negative
impact on the log of growth. A proportionate (appropriately signed), its absolute value lies
change in capital stock and electricity between zero and one, and its statistically
consumption will bring about a lesser significant. This implies a long-run convergence
proportionate change in growth while a of the model; it hereby implies that if any
proportionate change in labour force will bring external shock is introduced into the model, the
about a more proportionate change in growth. model would still converge with time. The
The result obtained from the estimation was speed of error adjustment of the model is quite
found consistent to that of Wolde-Rufael Y impressive (about 99%), implying 99% of
(2004), Akinlo A.E (2009) and Kauakou A.K present error in the model would be corrected
which also found a positive cointegrated and in the long-run
significant impact of electricity consumption on
the level of economic growth.

Variable D(LGDP) D(LKAP) D(LLAB) D(LELEC)


ECT_1 -0.985996 7.746074 -0.004681 -2.311518
(0.32424) (8.75592) (0.16940) (2.40066)
[-3.04091] [0.88467] [-0.02763] [-0.96287]

Causality Test directional relationship between electricity


consumption and GDP for the observed period,
The causality test using the pairwise approach implying that as the level of electricity
shows the causal relationship between consumption increases; the growth of the
electricity consumption and GDP with f-stat of Nigerian economy is enhanced and vice versa.
3.41182 and probability of 0.05040, due to the The obtained result is similar to the works of
significance of the probability; we hereby Odhiambo N.M (2010), Ouedraogo N.S (2012)
conclude that electricity consumption does and Akinlo A.E (2009) which all concluded that
granger cause GDP for the observed period. increasing demand for energy is an engine of
Also, the result indicates that GDP does granger development for developing countries.
cause electricity consumption. This implies bi-

Pairwise Granger Causality Test


Null Hypothesis F-statistics Prob
LELEC does not Granger Cause LGDP 3.41182 0.05040
LGDP does not Granger Cause LELEC 4.64951 0.02015

Source: Computed by author using e-views 5.


0

7
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

Conclusion differencing mechanism at first-order


integration. The study found the existence of a
This paper attempts to examine the relationship unique co-integrating relationship among the
that exists between electricity consumption and variables in the model, as well the VECM
economic growth in Nigeria using the Johansen estimates indicates a possibility of a long run
and Juselius co-integration technique of convergence with high speed of error
estimation based on Cobb-Douglas growth correction. The indicator of electricity
model for the period covering 1980-2010. The consumption was found to exert a very
study used the electricity consumption data significant impact on growth. In line with the
readily available from WDI as against that obtained result, there exist a bi-directional
provided by the Central Bank of Nigeria causal relationship between electricity
Statistical Bulletin, since the latter exhibited consumption and economic growth. The
some form of inconsistencies from time to time. inelastic impact of electricity consumption on
The study conducted a unit root testing to growth, as obtained in the analysis; therefore
ascertain the stationery status of the data series; call for the need to strengthen the effectiveness
as theories as proofed the non-stationary of of energy generating agencies by ensuring
most economic data in level state. The series periodic replacement of worn-out equipment
were found to contain unit root, hereby and necessary tools in order to drastically
necessitating the incorporation of the reduce power losses.

References

[1] Akinlo, A.E.(2009). Electricity Consumption and Economic growth in Nigeria: Evidence from
Cointegration and Co-feature Analysis. Journal of Policy Modeling 31(5), 681-693
[2] Ahmad N; Hayat M.F: Hamad N & Iugman M. (2012). Energy Consumption and Economic
growth: Evidence from Pakistan. Australian Journal of Business and Management Research. Vol.
2 No. 06 (09-14)
[3] Altmay, G, and Karagol E, (2005). Electricity Consumption and Economic Growth: Evidence for
Turkey. Energy Economies 27(6), 859-856.
[4] Asafu-Adjaye, J (2000). The Relationship Between Energy Consumption, Energy Prices and
Economic Growth: Time Series Evidence from Asian Developing Countries. Energy Economies
22(6) 615-625
[5] Ciarreta A. and Zarraga A. (2007). Electricity Consumption and Growth: Evidence from Spain.
Department of Economic analysis II, University of the Basque Country, Lehendakari Agirre 83,
Bilbao 48015, Spain.
[6] Ghosh, S. (2002). Electricity Consumption and Economic Growth in India. Energy policy 30(2),
125-129
[7] Johanseh, S. and Juselius K (1990). Maximum Likelihood Estimation and Inference on Co-
integrationWith Applications to The Demand for Money. Oxford Bulletin of Economics and
Statistics 52, 162-210.
[8] Kouakou K. Auguste (2010). Economic Growth and Electricity Consumption in Cote DIvoire.
Evidence from time series analysis. Energy policy. Science Direct.

[9] Kraft, J and Kraft A. (1978). On the Relationship Between Energy and GDP. Journal of Energy and
Development 3, 401-403
[10] Morimoto R. and Hope C. (2001). The impact of Electricity Supply on Economic Growth in Sri
Lanka.The Judge Institute of Management Studies, Trumpington Street, Cambridge CB21AG.
[11] Quedraogo, I.M. (2010). Electricity Consumption and Economic Growth in Burkina Faso: a co-
integration Analysis. Energy Economies 32(3), 524-531.

8
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

[12] Wolde-Rufael Y. (2009). Energy Consumption and Economic Growth: the Experience of African
Countries revisited. Energy policy 34(10), 1106-1114.
[13] Yang, H.Y (2000). A note on the causal relationship between energy and GDP in Taiwan. Energy
Economies 22(3), 309-317.
[14] Yoo, S. (2005). Electricity Consumption and Economic growth: Evidence from Korea. Energy
policy 33(12), 1627-1632.

APPENDIX

Date: 01/24/13 Time: 23:31


Sample (adjusted): 1992 2008
Included observations: 17 after adjustments
Trend assumption: Quadratic deterministic trend
Series: LGDP LKAP LLAB LELEC
Lags interval (in first differences): 1 to 1

Unrestricted Co-integration Rank Test (Trace)

Hypothesized Trace 0.05


No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.975426 97.51483 55.24578 0.0000


At most 1 0.667875 34.51174 35.01090 0.0565
At most 2 0.469386 15.77360 18.39771 0.1122
At most 3 * 0.254827 5.000350 3.841466 0.0253

Trace test indicates 1 co-integrating eqn(s) at the 0.05 level


* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Co-integration Rank Test (Maximum Eigenvalue)

Hypothesized Max-Eigen 0.05


No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.975426 63.00309 30.81507 0.0000


At most 1 0.667875 18.73815 24.25202 0.2267
At most 2 0.469386 10.77325 17.14769 0.3299

9
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

At most 3 * 0.254827 5.000350 3.841466 0.0253

Max-eigenvalue test indicates 1 co-integrating eqn(s) at the 0.05 level


* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Co-integrating Coefficients (normalized by b'*S11*b=I):

LGDP LKAP LLAB LELEC


-73.57378 5.436044 -244.6264 9.267434
-12.43977 -1.993485 81.79619 11.23605
8.557411 -2.120117 -80.61079 -0.793924
6.085605 -1.551089 74.07431 -5.524506

Unrestricted Adjustment Coefficients (alpha):

D(LGDP) 0.013401 0.010721 0.002844 -0.000303


D(LKAP) -0.105283 0.142125 0.176942 -0.106376
D(LLAB) 6.36E-05 -0.000900 0.002050 -0.003304
D(LELEC) 0.031418 -0.039546 0.052314 0.024985

1 Co-integrating Equation(s): Log likelihood 146.2102

Normalized co-integrating coefficients (standard error in parentheses)


LGDP LKAP LLAB LELEC
1.000000 -0.073886 3.324912 -0.125961
(0.00235) (0.11649) (0.00723)

Adjustment coefficients (standard error in parentheses)


D(LGDP) -0.985996
(0.32424)
D(LKAP) 7.746074
(8.75592)
D(LLAB) -0.004681
(0.16940)
D(LELEC) -2.311518
(2.40066)

10
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

2 Co-integrating Equation(s): Log likelihood 155.5793

Normalized co-integrating coefficients (standard error in parentheses)


LGDP LKAP LLAB LELEC
1.000000 0.000000 0.200714 -0.371243
(0.87773) (0.04686)
0.000000 1.000000 -42.28426 -3.319752
(11.8478) (0.63257)

Adjustment coefficients (standard error in parentheses)


D(LGDP) -1.119364 0.051479
(0.21010) (0.01630)
D(LKAP) 5.978075 -0.855647
(8.22260) (0.63804)
D(LLAB) 0.006520 0.002141
(0.17049) (0.01323)
D(LELEC) -1.819573 0.249622
(2.24881) (0.17450)

3 Co-integrating Equation(s): Log likelihood 160.9659

Normalized co-integrating coefficients (standard error in parentheses)


LGDP LKAP LLAB LELEC
1.000000 0.000000 0.000000 -0.376676
(0.04603)
0.000000 1.000000 0.000000 -2.175137
(0.75819)
0.000000 0.000000 1.000000 0.027070
(0.01633)

Adjustment coefficients (standard error in parentheses)


D(LGDP) -1.095025 0.045448 -2.630685
(0.20040) (0.01645) (0.72104)
D(LKAP) 7.492240 -1.230785 23.11685
(7.13014) (0.58536) (25.6549)
D(LLAB) 0.024063 -0.002206 -0.254469
(0.16455) (0.01351) (0.59207)

11
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

D(LELEC) -1.371897 0.138710 -15.13742


(1.89204) (0.15533) (6.80773)

Vector Error Correction Estimates


Date: 01/24/13 Time: 23:42
Sample (adjusted): 1992 2008
Included observations: 17 after adjustments
Standard errors in ( ) & t-statistics in [ ]

Co-integrating Eq: CointEq1

LGDP(-1) 1.000000

LKAP(-1) -0.073886
(0.00235)
[-31.4478]

LLAB(-1) 3.324912
(0.11649)
[ 28.5426]

LELEC(-1) -0.125961
(0.00723)
[-17.4143]

@TREND(80) -0.123061

C -75.86021

Error Correction: D(LGDP) D(LKAP) D(LLAB) D(LELEC)

CointEq1 -0.985996 7.746074 -0.004681 -2.311518


(0.32424) (8.75592) (0.16940) (2.40066)
[-3.04091] [ 0.88467] [-0.02763] [-0.96287]

D(LGDP(-1)) 0.367783 -3.952285 0.049494 1.933972


(0.22320) (6.02726) (0.11661) (1.65253)

12
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

[ 1.64779] [-0.65574] [ 0.42445] [ 1.17031]

D(LKAP(-1)) -0.027700 0.135965 0.001435 -0.118703


(0.01790) (0.48344) (0.00935) (0.13255)
[-1.54726] [ 0.28125] [ 0.15340] [-0.89556]

D(LLAB(-1)) 0.948793 -7.390596 -0.117253 5.185533


(0.81096) (21.8993) (0.42368) (6.00424)
[ 1.16996] [-0.33748] [-0.27675] [ 0.86365]

D(LELEC(-1)) -0.061404 1.525942 0.025138 -0.607433


(0.05420) (1.46366) (0.02832) (0.40130)
[-1.13289] [ 1.04255] [ 0.88772] [-1.51366]

C -0.070296 -0.474103 0.035394 -0.256863


(0.03591) (0.96983) (0.01876) (0.26590)
[-1.95734] [-0.48885] [ 1.88635] [-0.96600]

@TREND(80) 0.003742 0.032546 -0.000507 0.006106


(0.00121) (0.03259) (0.00063) (0.00893)
[ 3.10123] [ 0.99878] [-0.80385] [ 0.68339]

R-squared 0.745612 0.272122 0.169489 0.258426


Adj. R-squared 0.592979 -0.164605 -0.328817 -0.186518
Sum sq. resids 0.003302 2.407722 0.000901 0.180994
S.E. equation 0.018171 0.490685 0.009493 0.134534
F-statistic 4.885001 0.623094 0.340131 0.580806
Log likelihood 48.52341 -7.508430 59.56031 14.48936
Akaike AIC -4.885107 1.706874 -6.183565 -0.881101
Schwarz SC -4.542020 2.049962 -5.840478 -0.538013
Mean dependent 0.041489 -0.090937 0.025718 0.045978
S.D. dependent 0.028482 0.454688 0.008235 0.123508

Determinant resid covariance (dof adj.) 3.32E-12


Determinant resid covariance 3.98E-13
Log likelihood 146.2102
Akaike information criterion -13.43650
Schwarz criterion -11.86810

13
Vol. II, Issue 4
Electricity Consumption and Economic Growth in Nigeria
July 2013

Pairwise Granger Causality Tests


Date: 01/25/13 Time: 00:23
Sample: 1980 2010
Lags: 2

Null Hypothesis: Obs F-Statistic Probability

LKAP does not Granger Cause LGDP 17 1.06842 0.37409


LGDP does not Granger Cause LKAP 0.38695 0.68730

LLAB does not Granger Cause LGDP 19 1.37442 0.28510


LGDP does not Granger Cause LLAB 1.39448 0.28037

LELEC does not Granger Cause LGDP 28 3.41182 0.05040


LGDP does not Granger Cause LELEC 4.64951 0.02015

LLAB does not Granger Cause LKAP 17 0.05402 0.94764


LKAP does not Granger Cause LLAB 0.04386 0.95724

LELEC does not Granger Cause LKAP 17 0.93410 0.41973


LKAP does not Granger Cause LELEC 1.67644 0.22801

LELEC does not Granger Cause LLAB 18 2.57734 0.11408


LLAB does not Granger Cause LELEC 0.72138 0.50455

14